IGNITE: Practical AI with Machine Learning for Observability
summary
Netdata runs distributed edge agents that keep all data locally and emit thousands of per-second time series, then trains 18 ML models per metric to model recent behaviour. Anomalies are noisy individually but cluster across metrics and across servers, so Netdata uses a scoring engine to order signals and surfaces an anomaly rate ribbon on charts. Dashboards include a needle framework to filter by node, label or component and produce ordered lists for fast root-cause triage.
AI-written summary · about
related talks
22mAI with Machine Learning for Observability in Netdata
44mNetdata: Open Source, Distributed Observability Pipeline – Journey and Challenge
42mMachine Learning In Observability Systems: From Insights To Action
29mAI Driven Observability based on Open Source
01h 00mWhy observability requires the marriage of AI, metrics, and logs
25mMonitoring AI with AI
30mBetter data quality with ML & visualization
17mSponsored keynote: Can AI handle the complexity of Kubernetes and cloud-native environments?